UPGRADE, Vol. XI, issue no. 3, June 2010 - Council of European ...

UPGRADE is the European Journal for the Informatics Professional, published bimonthly at Publisher UPGRADE is published on behalf of CEPIS (Council of European Professional Informatics Societies, ) by Novática , journal of the Spanish CEPIS society ATI (Asociación de Técnicos de Informática, )

Vol. XI, issue No. 3, June 2010

UPGRADE monographs are also published in Spanish (full version printed; summary, abstracts and some articles online) by Novática UPGRADE was created in October 2000 by CEPIS and was first published by Novática and INFORMATIK/INFORMATIQUE, bimonthly journal of SVI/FSI (Swiss Federation of Professional Informatics Societies, ) UPGRADE is the anchor point for UPENET (UPGRADE European NETwork), the network of CEPIS member societies’ publications, that currently includes the following ones: • inforewiew, magazine from the Serbian CEPIS society JISA • Informatica, journal from the Slovenian CEPIS society SDI • Informatik-Spektrum, journal published by Springer Verlag on behalf of the CEPIS societies GI, Germany, and SI, Switzerland • ITNOW, magazine published by Oxford University Press on behalf of the British CEPIS society BCS • Mondo Digitale, digital journal from the Italian CEPIS society AICA • Novática, journal from the Spanish CEPIS society ATI • OCG Journal, journal from the Austrian CEPIS society OCG • Pliroforiki, journal from the Cyprus CEPIS society CCS • Tölvumál, journal from the Icelandic CEPIS society ISIP Editorial TeamEditorial Team Chief Editor: Llorenç Pagés-Casas Deputy Chief Editor: Rafael Fernández Calvo Associate Editor: Fiona Fanning Editorial Board Prof. Vasile Baltac, CEPIS President Prof. Wolffried Stucky, CEPIS Former President Hans A. Frederik, CEPIS Vice President Prof. Nello Scarabottolo, CEPIS Honorary Treasurer Fernando Piera Gómez and Llorenç Pagés-Casas, ATI (Spain) François Louis Nicolet, SI (Switzerland) Roberto Carniel, ALSI – Tecnoteca (Italy) UPENET Advisory Board Dubravka Dukic (inforeview, Serbia) Matjaz Gams (Informatica, Slovenia) Hermann Engesser (Informatik-Spektrum, Germany and Switzerland) Brian Runciman (ITNOW, United Kingdom) Franco Filippazzi (Mondo Digitale, Italy) Llorenç Pagés-Casas (Novática, Spain) Veith Risak (OCG Journal, Austria) Panicos Masouras (Pliroforiki, Cyprus) Thorvardur Kári Ólafsson (Tölvumál, Iceland) Rafael Fernández Calvo (Coordination) English Language Editors: Mike Andersson, David Cash, Arthur Cook, Tracey Darch, Laura Davies, Nick Dunn, Rodney Fennemore, Hilary Green, Roger Harris, Jim Holder, Pat Moody. Cover page designed by Concha Arias-Pérez "DNA in love" / © CEPIS 2010 Layout Design: François Louis Nicolet Composition: Jorge Llácer-Gil de Ramales Editorial correspondence: Llorenç Pagés-Casas Advertising correspondence: UPGRADE Newslist available at Copyright © Novática 2010 (for the monograph) © CEPIS 2010 (for the sections UPENET and CEPIS News) All rights reserved under otherwise stated. Abstracting is permitted with credit to the source. For copying, reprint, or republication permission, contact the Editorial Team

Monograph: 2010 - Emerging Information Technologies (I) (published jointly with Novática*)

Guest Editors: Alonso Álvarez-García, Heinz Brüggemann, Víctor-Amadeo Bañuls-

Silvera, and Gregorio Martín-Quetglas 3 Presentation: The Future is getting Closer — Alonso ÁlvarezGarcía, Heinz Brüggemann, Víctor-Amadeo Bañuls-Silvera, and Gregorio Martín-Quetglas 7 The Challenge of Future Communications — José-Luis NúñezDíaz and Óscar-Miguel Solá 13 Building the Future Telecommunications: Services and Networks of Internet — Heinz Brüggemann, Jukka Salo, José Jiménez, and Jacques Magen 20 Engineering Future Network Governance — Ranganai Chaparadza, Martin Vigeraux, José-Antonio Lozano-López, and Juan-Manuel González-Muñoz 30 Key Factors for the Adoption of Cloud Technologies by Telco Operators — Juan-Antonio Cáceres-Expósito, Juan-José HierroSureda, Luis M. Vaquero-González, and Fernando de la Iglesia-Medina 33 Trends in Natural Language Processing and Text Mining — Javier Pueyo and José-Antonio Quiles-Follana 40 Security 2.0: Facing up to the Tsunami — Enrique Díaz-Fernández, Miguel Ochoa-Fuentes, David Prieto-Marqués, Francisco RomeroBueno, and Vicente Segura-Gualde 46 Trust in the Information Society: RISEPTIS Report — RISEPTIS, Advisory Board of the Think-Trust Project

UPENET (UPGRADE European NETwork) 53 From Mondo Digitale (AICA, Italy) Green Computing

Green Software — Giovanna Sissa

CEPIS NEWS 64 Selected CEPIS News — Fiona Fanning

The opinions expressed by the authors are their exclusive responsibility ISSN 1684-5285

Monograph of next issue (August 2010) "2010: Emerging Information Technologies (II)" (The full schedule of UPGRADE is available at our website)

* This monograph will be also published in Spanish (full version printed; summary, abstracts, and some articles online) by Novática, journal of the Spanish CEPIS society ATI (Asociación de Técnicos de Informática) at .

UPENET

Green Computing

Green Software Giovanna Sissa

© Mondo Digitale, 2009 This paper was first published, in its original Italian version, under the title "Green Software", by Mondo Digitale (issue no. 3, September 2009, pp. 46-54, available at ). Mondo Digitale, a founding member of UPENET, is the digital journal of the CEPIS Italian society AICA (Associazione Italiana per l’Informatica ed il Calcolo Automatico, .)

An effective insight about ICT environmental sustainability requires to pay attention also to the software features, being this another responsible for the CO2 emissions. The energy consumption in the computer’s use phase doesn’t depend only from hardware but also from software configuration and from its efficiency. Software is also responsible for the Induced hardware obsolescence. The actual computer lifecycle is shorter than the potential one. Some real data and interview will give evidence on how software improvements, aimed at making the information system configuration more effective, will lead to energy consumption reduction. A software-based approach will also allow a longer use for PCs, respecting the environment, saving energy, emissions and money, and, in the meantime, moving toward the cloud computing paradigm. This paper includes an interview with Gianluigi Castelli, Executive Vice President Information & Communication Technology of the Italian company ENI.

Keywords: Environmental Sustainability, eWaste, Green Computing, Green Software, IT Management Software, Thin Client.

1 The Environmental Impact of the ICT Sector Information and communication technologies (ICTs) have been contributing to environmental problems: computers, electronic devices and ICT infrastructure consume significant amounts of electricity, placing a heavy burden on our electric grids and contributing to greenhouse gas emissions (see Figure 1). In 2007, the total footprint of the ICT sector – including personal computers (PCs) and peripherals, telecom networks and devices and data centers – was 830 MtCO2 emissions, about 2% of the estimated total emissions from human activity released that year (a figure equivalent to aviation industry) [1]. © CEPIS

Additionally, ICT hardware poses severe environmental problems both during its production and its disposal. Each stage of a computer’s life cycle, from its production, throughout its use, and into its disposal, presents environmental problems. Manufacturing computers and their various electronic and non-electronic components consumes electricity, raw materials, chemicals, and water, and generates hazardous waste [2,3]. Use phase is also energy intensive. All these, directly or indirectly, increase carbon dioxide emissions and impact the environment. The trend is to increase in the BAU (Business As Usual) scenario [1] (Figure 1). The total electrical energy consumption by servers, computers, monitors, data communications equipment, and cooling systems for data centers is steadily increasing. The effects of ICT on the environ-

Author Giovanna Sissa is engaged on research on ICT since early ‘80s. She has been consultant for several large companies and, as an independent expert, she collaborates with several national and international institutions. She was the head of the Osservatorio Tecnologico of the Italian Ministry of Education till 2009. She was appointed member of the Ministry of Public Administration’s Open Source Commission and member of the Register of Experts in Innovation of the Ministry of Industry. She is author of books and papers on emerging ICT issues. Trying to relate the two topics of ICT and environment, she published, in 2008, "Il computer sostenibile" (The Sustainable Computer), an analysis on the environmental impact of computers, highlighting the role that software, especially open source software, can play to respect the environment.

UPGRADE Vol. XI, No. 3, June 2010 53

UPENET ment are commonly ranked in first, second and third order effects [3,4]. The first order effects are directly related to the mere physical existence of ICT and include production, use and end-of-life treatment. The second order effects are related to the application of ICT and include effects leading to optimization of processes in other sectors (e.g. traffic optimization), substitution effects (e.g. e-processes that replace traffic) and induction effects (when ICT creates more demand in other sectors). The third order effects are related to the societal changes that ICT brings along and they are more difficult to foresee and quantify. This includes the deep structural change towards a de-materialized economy, the rebound effects, and the increased dependency on the critical infrastructure of ICT.

1.1 A Rebound Effect Each PC in use generates about a ton of carbon dioxide every year [5]. ICT equipment accounts for about 10% of the UK’s total electricity consumption (worth the equivalent of four nuclear power stations). Between 2000 and 2006, energy consumption from non-domestic ICT equipment increased by more than 70% and it is expected to grow by a further 40% by 2020 [6]. The energy consumption resulting from the use of computers, the internet and other forms of ICTs has risen enormously over recent years [9]. ICTs are continuously making astounding progress in technical efficiency. The time, space, material and energy needed to provide a unit of ICT service have decreased by three orders of magnitude (a factor of 1000) since the first PC was sold [4]. ICT devices are becoming increasingly more compact and energy efficient. Computers are continuously mak-

ing astonishing progress in energy efficiency, measured in performance per watt1 , due to innovative design techniques, coming from technological aspects to the processing architectural dynamic management. The power density2 is also increasing [8]. The well-known principle called Moore’s Law, according to which the number of transistors per microchip doubles every 18-24 months, has yet to be disproved. But Moore‘s Law in physical terms and Moore‘s Law in economic terms are different. For an exponential growth of performance, we have a double exponential growth per chip of performance per money. The miniaturization paradox indicates that hardware is getting cheaper faster than it is getting smaller [3, 9]. By the same token, though, computers are getting cheaper – and so their use is therefore becoming more widespread. The same is true for using internet and telecommunication services. More and more powerful devices are used by more and more people. The number of devices and uses is currently rising faster than corresponding developments in energy and materials efficiency. Research on energy and sustainability refers to this phenomenon as a "rebound effect" [3]. Rebound effects occur if, and when, the efficiency of providing a service is increased but there is no factor limiting the demand for this service, such as the price to be paid or the time needed for consuming it. The economic system (as a functional system of society) adapts to the higher efficiency level at which the service is provided increasing the demand for the service [4]. The demand for ICT is increasing even faster than the energy efficiency of ICT devices; the total energy demand of the installed hardware base is growing [5,10]. As manufacturers competed to create ever-faster processors, smaller and

1 Performance per watt is a measure of the energy efficiency of a particular computer architecture. Literally it measures the rate of computation that can be delivered by a computer for every watt of power consumed. The performance and power metrics used depend on the definition; reasonable measures of performances are FLOPS, MIPS, or the score for any performance benchmark 2 Power density can be measured in W/m2 or W/ft2 (Watts per square meter of foot).

54

UPGRADE Vol. XI, No. 3, June 2010

smaller transistors (running hotter and consuming more electricity) were used to form the basis of each new generation. Increased operating temperature added to the consumption of power, requiring more and more cooling fans. Modern IT systems provide more computing power per unit of energy (kWh) and thus reduce energy consumption per unit of computing power. Despite this, they are actually responsible for an overall increase in energy consumption and for an increase in the cost of energy as a proportion of IT costs. This is because users are not simply using the same amount of computing power as before, while using the new technology to reduce their power consumption (or operating temperatures), nor are they using technology to leverage savings in energy costs or in CO2 production. Instead, users are taking and using the increased computing power offered by modern systems regardless implication on sustainability. New software in particular is devouring more and more power every year. Some software requires almost constant access to the hard drive, draining power much more rapidly than previous packages did. The advent of faster, smaller chips has also allowed manufacturers to produce smaller, stackable and rackable servers allowing greater computing power to be brought to bear but with no reduction in overall energy consumption, and often with a much greater requirement for cooling [11].

2 Green ICT and Green Software Green ICT, or Green Computing, is the study and practice of designing, manufacturing, using, and disposing of computers, servers, and associated subsystems - such as monitors, printers, storage devices, and networking and communications systems - efficiently and effectively with minimal or no impact on the environment [7]. Green ICT includes the dimensions of environmental sustainability, the economics of energy efficiency, and the total cost of ownership, which includes the cost of disposal and recycling. Green ICT benefits the environment by improving energy efficiency, lowering © CEPIS

UPENET

% of 830 MtCO2 emissions

37%

Telecoms infrastructures and devices DATA centres

49%

14%

PC, peripheralsand printers

Figure 1: ICT Footprint by Sector 2007 (Gesi Group, Smart 2020).

greenhouse gas emissions, using less harmful materials, and encouraging reuse and recycling [5]. Green design, green manufacturing, green use, green disposal are complementary paths of green ICT. Only focusing on these four fronts we can achieve total environmental sustainability from the IT side and make IT greener throughout its entire lifecycle. Green design and manufacturing are not the matter of this paper. The focus will be on the green use and also on green disposal, and on the role played by the software in a green strategy. Corporate and their IT department are recognizing their impact on "carbon footprint" on the environment and are actively engaged in finding green IT solutions. While many have looked to hardware and power systems as part of these solutions, it is becoming increasingly recognized that software management is also a key part in the role of the ICT sector [12]. It is possible to tune a piece of software to do the job it was doing before, but on less power and less CO2 generation (not exactly the same), or even write software from scratch to be ‘greener’ and more energy-efficient. This goal is easier to achieve on hardware that is energy-efficient already

3 .

© CEPIS

such as smart handhelds, laptops and newer desktops/servers designed and built after it became fashionable to compete on computing power per watt 3 . The micro-processors within electronic equipment require energy both to operate and for cooling fans. Improvements in chip design (such as ‘multi-core’ processors) can save 3060% of the energy used by the processor if software is written to take advantage of this capacity [10]. Low power software, used since time in the design of embedded systems, deals with design oriented toward energy saving, starting from low level compiling and programming aspects (like, in example, data architecture) up to operating systems. The growth of ICT leads to a larger attention in energy saving in all sectors. The continued rise of Internet and Web applications is driving the rapid growth of data centers. Enterprises are installing more servers or expanding their capacity. The number of server computers in data centers has increased six fold up to 30 million in the last decade, and each server draws far more electricity than earlier models [5,11,6], because is more powerful. A key green objective in using computer systems and operating data centers is to reduce their energy consumption, thereby minimizing the greenhouse gas emissions [5]. But green ICT is not only a matter of hardware. Software plays a critical role on each phase of the life cycle. In

the use phase, as we will see later, in an interview where will be given data and evidence about the energy saving coming from better software efficiency. And software can play a green role also at the end of the hardware lifespan. Green computing can be aided by software solutions as well. There are lots of software-based solutions that not only monitor and better utilize existing resources, but can also go beyond and nullify the need to add extra hardware as the company expands, meeting the green requirement with current infrastructure. I will use the term "green software" in a broader sense than the usual meaning of it. I define as "green" every software taking into account some environmental aspects and allowing ICT to be greener. Let’s see how, starting from the Data Center, software is put into relation with energy consumption.

2.1 Greening Data Centers It is estimated that a medium-sized server has roughly the same annual carbon footprint as a SUV vehicle running 15 miles per gallon. The power required for a rack of high density server blades can be 10-15 times greater than a traditional server [6]. To understand how to reduce energy consumption in a data center, we need to know where and how energy is used. Each component is divided into two portions: IT equipment (servers, storage, and network) uses about 45% of the energy, and the infrastructure

UPGRADE Vol. XI, No. 3, June 2010 55

UPENET that supports this equipment - such as chillers, humidifiers, computer room air conditioners, power distribution units (PDU), uninterruptable power supplies (UPS), lights, and power distribution -uses the other 55% of the energy[13]. Thus, IT equipment does not use 55% of the energy that is brought into the data center, so this portion of the energy is not producing calculations, data storage, and so forth. Energy savings and efficiencies for non-IT equipment can reduce this inefficiency [13]. As far as the energy consumption of the IT equipment level, in a typical server, the processor uses only 30% of the energy and the remaining of the system uses 70%. Commonly, servers are underutilized, yet they consume almost the same amount of energy as though they were running at 100%. A typical server utilization rate is 20%. Underutilized systems can be a big issue because a lot of energy is expended on non-business purposes, thus wasting a major investment [13]. For each watt used for computational resources, the processor uses 5 watts, the server 16 watts and the data center 27 watts, i.e. we need to spend 27 watts of power for each watt spent for resource usage. Data center efficiency can be improved by using new energy-efficient equipment, improving airflow management to reduce cooling requirements, adopting environmentally friendly designs for data centers, investing in energy management software. The measure of direct electricity consumption to power servers compared to the indirect electricity used to cool the equipment is known as the Power Usage Effectiveness (PUE). As PUE increases, the indirect cooling electricity consumption, compared to the direct electricity used by the server increases. A PUE value of less than two is considered to be good practice [6,14, 15]. The European Commission launched a "voluntary code of conduct for data centres" in November 2008, which aims to raise awareness of their energy use. It recommends best practice and targets for data centre owners, whilst avoiding prescription of specific 56

UPGRADE Vol. XI, No. 3, June 2010

technologies [14,10]. Software solutions are very important to energy efficiency: virtualisation, efficiency, data center consolidation, power server management are important areas for energy efficiency [12].

2.2 Virtualisation The biggest technology that has enabled software-centric green ICT approach is virtualization. The idea is pretty simple: a host software (a control program), creates a simulated computer environment (a virtual machine) for its guest software. The guest software, which is often itself a complete operating system, runs just as if it were installed on a stand-alone hardware platform. With virtualization, one physical server hosts multiple virtual servers. Virtualization enables data centers to consolidate their physical server infrastructure by hosting multiple virtual servers on a smaller number of more powerful servers, using less electricity and simplifying the data center. Besides getting better hardware usage, virtualization reduces data center floor space, makes better use of computing power, and reduces the data center’s energy demands [5]. Virtualisation technology allows fewer servers to store the same amount of data. Server virtualisation involves a software application dividing one physical server into multiple isolated virtual environments [6]. Virtualisation and server consolidation can allow users to ‘do more with less’, allowing one large server to replace several smaller machines. This can reduce the power required and the overall heat produced. By reducing the number of servers in use, users can simplify their IT infrastructure, and reduce the power and cooling requirements [11]. 2.3 Software and Energy Efficiency Despite the trend towards server virtualisation and consolidation in some companies, business demand for IT services is increasing, and many companies are still expanding their data centers, while the number of servers in such data centers is still increasing

annually by about 18% [11]. Several measures can be taken in order to increase energy efficiency in data centers. The cause-effect chain starts with applications and continues through the IT hardware and the power supply, right up to building planning and cooling as well as energy management. The key idea is that measures are most effective at the start of this chain. It can look trivial to say, but when an application is no longer needed and the accompanying server is therefore switched off, less power is used, the losses in the UPS (Uninterruptible Power Supply) system decline, and the cooling load is reduced [7]. The starting points for an energy efficient data center are the applications and the data [7]. It often happens that, although a third of all applications is no longer needed, they nevertheless continue to operate on the server. And the difference in the extent to which certain largely identical applications use hardware resources is often not insignificant. For the management of the energy efficient data center, this means that the applications and the data must be examined regularly, and that applications and data which are no longer needed should be deleted as far as possible.

2.4 De-duplication of Data We live in an era of unprecedented information growth. The Moore Law and the idea that storages resource costs are decreasing endlessly, have as negative side effect that data managers don’t pay sufficient attention to the redundancy of data. In 2006, 161 exabytes (161x1012 bytes) of digital information was created and copied. That is equivalent to three million times the information in all the books ever written, or 12 stacks of books, each extending from the Earth to the sun [16] It is estimated that the amount of information created and copied in 2011 will increase six-fold from 2006 [16]. Data de-duplication essentially refers to the elimination of redundant data. In the deduplication process, duplicate data is deleted, leaving only one copy of the data to be stored. However, indexing of © CEPIS

UPENET Payroll values per employee (about 35.000) Before Optim.

After Optim.

Access to a view (time)

3.5 sec

0,01 sec

Access to a view (#buffer)

≈223.000

163

Power ratio